LEARNING AND ANALYZING VECTOR ENCODING OF SYMBOLIC REPRESENTATION
· · Paul Smolensky · Rishabh Singh
Abstract
We present a formal language with expressions denoting general symbol structures and queries which access information in those structures. A sequence-to-sequence network processing this language learns to encode symbol structures and query them. The learned representation (approximately) shares a simple linearity property with theoretical techniques for performing this task.
Video
Chat is not available.
Successful Page Load